CN113012757B - Method and system for identifying bases in nucleic acids - Google Patents

Abstract

The invention discloses a method for identifying bases in nucleic acids, a computer-readable storage medium, a computer program product and a system. The so-called method for identifying bases in nucleic acids includes mapping the coordinates of each bright spot in the bright spot set corresponding to the template onto the image to be detected, determining the position of the corresponding coordinates on the image to be detected; determining the position of the corresponding coordinates on the image to be detected; The intensity of the signal at the position of the corresponding coordinates, which is the corrected intensity; and comparing the intensity of the signal at the position of the corresponding coordinates on the image to be detected with the size of the first preset value, and judging the base corresponding to the position based on the comparison result. type to achieve base recognition. This method can quickly and accurately identify bases and achieve the determination of the nucleotide/base sequence of at least a part of the template sequence.

Description

Methods and systems for identifying bases in nucleic acids

Technical field

The present invention relates to the field of data processing, and in particular to a method for identifying bases in nucleic acids, a computer-readable storage medium, a computer program product and a system.

Background technique

In the related art, the so-called sequencing generally refers to the determination of biopolymers, including determination of the primary structure or sequence of nucleic acids such as DNA and RNA, including determination of the nucleotide bases (adenine A, adenine A, adenine, etc.) of a given nucleic acid fragment. The sequence of purine G, thymine T/uracil U and cytosine C). This type of method usually involves identifying the base at one or more positions in the nucleic acid, that is, performing base calling (basecalling), to determine the sequence of the nucleic acid.

The signal and/or the change in signal intensity corresponding to the binding of nucleotides/bases to a specific position of the nucleic acid molecule (template) to be tested can indicate the type of base at that position on the nucleic acid molecule. For example, different fluorescent molecules can be used to label it. Recognize different bases. The so-called nucleotide/base binding to a specific position of the nucleic acid molecule to be tested is also called the incorporation of nucleotides/bases into the nucleic acid molecule to be tested or base extension, for example, through polymerization, connection and hybridization. to fulfill.

Specifically, on a platform that uses an optical imaging system to collect images of base extension signals multiple times and implements nucleic acid sequencing based on processing these images, due to optical effects, spatial effects and/or chemical reactions such as chromatic aberration The effects of crosstalk and/or phasing on image acquisition, positioning and/or signal intensity often make it difficult to accurately identify bases based on image processing.

Therefore, how to process information including correlating images collected at multiple different time points to effectively and accurately determine the nucleotide/base type and sequence of at least a part of the template is a problem that is expected to be solved or improved.

Contents of the invention

The embodiments of the present invention are intended to solve one of the technical problems existing in the prior art to at least a certain extent or at least provide a useful means. To this end, embodiments of the present invention provide a method for identifying one or more bases in a nucleic acid, a computer-readable storage medium, a computer program product, and a system.

A method for identifying one or more bases in a nucleic acid according to an embodiment of the present invention, by detecting an image obtained from sequencing, including: mapping the coordinates of each bright spot in a bright spot set corresponding to a template to On the image to be inspected, determine the position of the corresponding coordinates on the image to be inspected; determine the intensity of the signal at the position of the corresponding coordinates on the image to be inspected, which intensity is the corrected intensity; and compare the signal of the position of the corresponding coordinates on the image to be inspected. Based on the comparison result between the intensity and the first preset value, the base type corresponding to the position is determined to realize base identification.

The so-called bright spot set corresponding to the template is constructed based on a set of images. Each image in the so-called set of images contains multiple bright spots; the so-called set of images and the image to be inspected are both from Sequencing and corresponding to the same field of view; the set of said images comes from at least one round of sequencing; at least a part of the so-called signals appear as at least a part of the so-called bright spots on the set of images.

Other embodiments of the invention relate to computer-readable media, computer products, computer program products and systems related to the methods of the above-described embodiments.

For example, a computer-readable storage medium according to an embodiment of the present invention is used to store a program for computer execution. Executing the program includes completing the method for identifying bases in nucleic acids in any of the above embodiments.

A computer product according to an embodiment of the present invention includes the computer-readable storage medium in any of the above embodiments.

A system according to an embodiment of the present invention includes the computer product in any of the above embodiments; and one or more processors for executing programs stored in so-called computer-readable storage media. Executing the program includes completing the base calling method in any of the above embodiments.

A computer program product according to an embodiment of the present invention includes instructions for identifying one or more bases in nucleic acids. When the computer executes the so-called program, the instructions cause the computer to execute the bases in any of the above embodiments. recognition methods.

An embodiment of the present invention is a system configured to perform the method for identifying bases in nucleic acids in any of the above embodiments.

A system according to an embodiment of the present invention includes a plurality of modules, which are used to perform the steps of the method for identifying bases in nucleic acids in any of the above embodiments.

The method, product and/or system for identifying bases in nucleic acids according to any embodiment of the present invention has no special restrictions on the type and format of the image to be inspected, that is, the original input data. The image to be inspected can come from any implementation based on optical imaging detection. Nucleic acid sequencing platforms include but are not limited to so-called second-generation and third-generation sequencing platforms, such as BGI including Complete Genomics, Illumina including PacBio (PacificBiosciences), and Thermo Fisher Scientific ThermoFisher includes one or more series of sequencing platforms from Life technologies, Roche and Helicos.

Using the method, product and/or system of any embodiment of the present invention to perform base identification can quickly and accurately identify bases and achieve the determination of the nucleotide/base sequence of at least a part of the template sequence.

Additional aspects and advantages of embodiments of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of embodiments of the invention.

Description of the drawings

The above and/or additional aspects and advantages of embodiments of the present invention will become apparent and readily understood from the description of the embodiments in conjunction with the following drawings, in which:

Figure 1 is a schematic flow chart of a method for identifying one or more bases in nucleic acids according to an embodiment of the present invention;

Figure 2 is the spectral curves of four fluorescent dyes according to the embodiment of the present invention;

Figure 3 is a 5*5 matrix schematic diagram of the embodiment of the present invention;

Figure 4 is a schematic diagram of a 5*5 size convolution kernel according to the embodiment of the present invention;

Figure 5 is a schematic comparison diagram of the image before and after convolution according to the embodiment of the present invention;

Figure 6 is a schematic diagram showing monotonic fluctuations in pixel values of multiple pixels in a specified direction of a 5*5 matrix according to an embodiment of the present invention;

Figure 7 is a schematic diagram of the process of dividing an image into multiple blocks and determining the offset between blocks to align a round of images in one field of view according to an embodiment of the present invention;

Figure 8 is a schematic diagram of the offset between at least part of the corresponding block combinations after dividing two images into blocks of 100*100 size according to an embodiment of the present invention;

Figure 9 is a schematic diagram of the construction process of a set of bright spots corresponding to the template for a set of images in one round of sequencing according to the embodiment of the present invention;

Figure 10 is a crosstalk scatter diagram between two images of four images in one field of view according to the embodiment of the present invention;

Figure 11 is an A-G crosstalk scatter diagram according to the embodiment of the present invention, with the abscissa being A and the ordinate being G;

Figure 12 is an A-T signal intensity fitting curve according to the embodiment of the present invention;

Figure 13 is a schematic diagram of the results of A-T signal intensity before and after correction according to the embodiment of the present invention;

Figure 14 is a schematic diagram of crosstalk between four images in one round of sequencing of one field of view after chromatic aberration correction according to the embodiment of the present invention;

Figure 15 is a schematic diagram of the signal crosstalk of the first and second rounds (cycle1 and cycle2) of a specific base in one field of view according to the embodiment of the present invention. From top to bottom and from left to right, they are A, C, G and The phasing scatter plot of T; in each phasing scatter plot, the abscissa is the relative signal intensity of the base in cycle1, and the ordinate is the relative signal intensity of the same base in cycle2;

Figure 16 is a schematic diagram of signal crosstalk in the thirtieth and thirty-first rounds (cycle30 and cycle31) of a specific base in one field of view according to the embodiment of the present invention. From top to bottom and from left to right, they are A and C. phasing scatter plot of , G and T; in each phasing scatter plot, the abscissa is the relative signal intensity of the base in cycle30, and the ordinate is the relative signal intensity of the same base in cycle31;

Figure 17 is a schematic diagram of the signal crosstalk of the sixtieth and sixty-first rounds (cycle60 and cycle61) of a specific base in one field of view according to an embodiment of the present invention. From top to bottom and from left to right, they are A and C. , G and T phasing scatter plot; in each phasing scatter plot, the abscissa is the relative signal intensity of the base in cycle60, and the ordinate is the relative signal intensity of the same base in cycle61;

Figure 18 is a schematic diagram of the signal crosstalk of the ninetieth and ninety-first rounds (cycle90 and cycle91) of a specific base in one field of view according to an embodiment of the present invention. From top to bottom and from left to right, they are A and C. , G and T phasing scatter plot; in each phasing scatter plot, the abscissa is the relative signal intensity of the base in cycle90, and the ordinate is the relative signal intensity of the same base in cycle91;

Figure 19 is a schematic diagram of the relationship between the phasing ratio or prephasing ratio of four bases and the number of sequencing rounds according to the embodiment of the present invention. The abscissa is the number of sequencing rounds, and the ordinate is the prephasing ratio;

Figure 20 is a phasing scatter plot of A in cycles 30 and 31 according to the embodiment of the present invention;

Figure 21 is a schematic diagram of the system 100 according to the embodiment of the present invention.

Detailed ways

Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The embodiments described below with reference to the drawings are exemplary and are only used to explain the present invention and are not to be construed as limitations of the present invention.

In the embodiments of the present invention, it should be understood that the terms "first" and "second" are only used for descriptive purposes and cannot be understood to indicate or imply relative importance or implicitly indicate the number of indicated technical features. ; Features defined as “first” and “second” may explicitly or implicitly include one or more of the described features. Unless otherwise stated, "a group" or "a plurality" refers to two or more.

It should be noted that, unless otherwise stated, "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection. Connection or mutual communication; it can be directly connected, or it can be indirectly connected through an intermediary, it can be the internal connection between two elements or the interactive relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in corresponding examples can be understood according to specific circumstances.

The present invention may repeat reference numbers and/or reference letters in different examples. Such repetition is for the purposes of simplicity and clarity and does not in itself indicate a relationship between the various embodiments and/or arrangements discussed.

In embodiments of the present invention, the terms "sequencing", "nucleic acid sequencing" and "gene sequencing" are interchangeable and refer to nucleic acid sequence determination; including sequencing by synthesis (sequencing by synthesis, SBS) and/or ligation sequencing ( Sequencing while ligating (SBL), including DNA sequencing and/or RNA sequencing, including long fragment sequencing and/or short fragment sequencing. The so-called long fragments and short fragments are relative, such as longer than 1Kb, 2Kb, 5Kb or 10Kb. Nucleic acid molecules can be called long fragments, and those shorter than 1Kb or 800bp can be called short fragments; including paired-end sequencing, single-end sequencing and/or paired-end sequencing, etc. The so-called paired-end sequencing or paired-end sequencing can refer to the same nucleic acid The reading of any two segments or parts of a molecule that do not completely overlap; the so-called sequencing includes the process of binding nucleotides (including nucleotide analogs) to a template and collecting the corresponding signals emitted.

Sequencing generally includes multiple rounds of sequencing to determine the order of multiple nucleotides/bases on the template; a "sequencing cycle" is also called a "sequencing round" and can be defined as four nucleotides/bases. One base extension of the base, in other words, can be defined as the completion of the determination of the base type at any specified position on the template. For sequencing platforms that implement sequencing based on polymerization or ligation reactions, one round of sequencing includes achieving four nucleosides at a time. The process of binding an acid (including nucleotide analogs) to a so-called template and collecting the corresponding signal emitted; for a platform based on polymerization to achieve sequencing, the reaction system includes the reaction substrate nucleotide, polymerase and template, and the template There is a sequence (sequencing primer) bound to it. Based on the principle of base pairing and polymerization reaction, the added reaction substrate nucleotide is catalyzed by the polymerase and connected to the sequencing primer to achieve the specific position of the nucleotide and the template. Combination; Generally, a round of sequencing can include one or more base extensions (repeat). For example, four nucleotides are added to the reaction system in sequence, and base extensions and corresponding reaction signals are collected respectively. A round of sequencing includes four base extensions; for another example, four nucleotides are added to the reaction system in any combination, such as two-two combinations or one-three combinations. The two combinations perform base extension and corresponding reaction signal collection respectively. One round of sequencing includes two base extensions; for another example, four nucleotides are added to the reaction system at the same time for base extension and reaction signal collection, and one round of sequencing includes one base extension.

The so-called "bright spots" (spots or peaks) on the image, also known as "bright spots" or "light spots", refer to the positions on the image where the signal is relatively strong. For example, the signal at this position is stronger than the surrounding ones. In the image It appears as a relatively bright spot or point, and a bright spot or a so-called position occupies one or more pixels. The bright spot/signal at this location may come from the target molecule or from non-target substances. Detection of "bright spots" involves detection of optical signals from target molecules such as extended bases or clusters of bases.

The so-called "chromatic aberration" (CA) refers to the optical phenomenon that the lens cannot focus all the colored light of various wavelengths on the same point [Max Born; Emil Wolf. Principles of Optics: Electromagnetic Theory of Propagation, Interference and Diffraction of Light(7th Edition).Cambridge University Press.October 13,1999:334.ISBN0521642221.]; In imaging, chromatic aberration manifests itself as the inability of each color on the spectrum to focus on the same point on the optical axis. For applications involving A sequencing platform that uses multiple wavelengths of colored light to image the same object (such as one or more nucleic acid molecules). At least, the chromatic aberration will cause the object to have different positions in multiple images of an object collected at different wavelengths. / coordinates, or in other words, the object does not actually move, but due to chromatic aberration it appears to be moving in multiple images at different wavelengths.

The so-called "crosstalk" (crosstalk or laser-crosstalk or spectrum-crosstalk), also known as "spectral crosstalk" or "spectral crosstalk", refers to the signal corresponding to one base spreading to the signal of another base. phenomenon; for sequencing platforms that use different labeled fluorescent molecules to identify different bases, if the emission spectra of two or more selected fluorescent molecules overlap, one fluorescent molecule may be detected in one round of sequencing The signal diffuses into another fluorescence channel.

The so-called "phase out of phase", "phase imbalance", "out of phase", and "phase difference" refer to the phenomenon of unsynchronized reactions between nucleic acid molecules in a group, such as a nucleic acid molecule cluster, in chemical reactions, including Lagging/lag (phasing or sequence lag) and advance/lead (prephasing or sequence lead); in a sequencing platform that uses different fluorescent molecules labeled to identify different bases, the fluorescent molecules corresponding to the bases at a specific position are expressed in The phenomenon of non-zero signal in more than one round of sequencing. Generally, nucleotides with fluorescent molecular labels and blocking groups are used for sequencing. The blocking group on the nucleotide can prevent other nucleotides from binding to the next position of the template. The blocking group is such as For the azide attached to the 3' position of the sugar group of the nucleotide, the loss of the blocking group or the failure to be removed before the extension of the next base will cause dephasing.

In the embodiment of the present invention, the images come from a platform that implements nucleic acid sequencing based on optical imaging detection chips. The so-called platforms include but are not limited to BGI/CG (Complete Genomics), Illumina/Solexa, ThermoFisher/Life Technologies/ABI SOLiD and One or more series of sequencing platforms from companies or institutions such as Roche 454.

In some platforms, multiple sequences (probes or sequencing primers) are fixed on a solid support such as a chip, and the template (nucleic acid molecule to be tested) is connected to the chip by binding to the probe, such as by hybridization. Optionally, The template is amplified on the chip, and then the chip carrying the template is loaded into the sequencing equipment. The sequencing equipment includes an imaging system and a liquid system. The liquid system is controlled to introduce solutions containing polymerase and nucleotides to The chip performs a controllable polymerase chain reaction under appropriate conditions. For example, the nucleotide solution passed through contains nucleotides including modified nucleotides, and the modified nucleotides have blocking groups and Fluorescent molecules, based on the principle of base complementarity, bind the modified nucleotide to a specific position of a template under the catalysis of polymerase, and the blocking group on it can prevent other nucleotides (including modified nucleosides) acid) is bound to the next position of the template; then, the imaging system is used to excite the fluorescent molecules so that the fluorescent molecules emit fluorescent signals, and these fluorescent signals are collected, such as taking pictures of the reaction area on the chip to obtain the image; finally, through the control liquid The cleavage reagent is passed through the system to remove the blocking group and fluorescent molecules of the modified nucleotide bound to the template; at this point, a base extension is completed, and the solution containing the polymerase and the nucleotide respectively is passed again to Chip, repeat the above base reaction. Based on the captured images and the time sequence of each photograph and/or the type of base added, the type of nucleotide/base bound to a specific position of the template each time is determined, that is, the specific positions of the template are determined. of nucleotides/bases.

The reaction efficiency based on individual steps of the biochemical reaction is less than 100%, such as the removal of modified nucleotides that are not bound to the template even before signal acquisition, such as the use of buffer pairs that do not affect base extension. The reaction area on the chip is cleaned. It is understandable that the positions that appear as bright spots on the collected images may not only correspond to the modified nucleotides bound to the template, but may also correspond to the nucleotides that are not bound to the template but cannot be removed. Modified nucleotides or fluorescent molecules may also correspond to signals emitted by other non-target substances in the detection area on the chip.

In one embodiment of the present invention, the image comes from a second-generation sequencing platform, such as the Illumina HiSeq/MiSeq series and the BGI MGISeq series. The input raw data is the position and intensity of the collected signal and other related parameters including the pixel related information of the image. The detection of so-called "bright spots" on the image involves the detection of optical signals corresponding to clusters of nucleic acid molecules.

Please refer to Figure 1, a method of identifying one or more bases in nucleic acids according to an embodiment of the present invention. The method detects images obtained from sequencing, including: S11 converting each bright spot set corresponding to the template The coordinates of the bright spot are mapped to the image to be inspected, and the position of the corresponding coordinates on the image to be inspected is determined; S21 determines the intensity of the signal at the position of the corresponding coordinates on the image to be inspected, and the intensity is the corrected intensity; and S31 compares The intensity of the signal at the position of the corresponding coordinate on the image to be detected is compared with the size of the first preset value, and the base type corresponding to the position is determined based on the comparison result to realize base identification.

The so-called bright spot set corresponding to the template is constructed based on a set of images. Each image in the set of images contains multiple bright spots; both the set of images and the image to be inspected are from sequencing and correspond to the same Field of view (FOV), the so-called set of images comes from at least one round of sequencing, and at least a part of the so-called signals appear as at least a part of the so-called bright spots on the set of images.

This method can quickly and accurately identify bases, and thereby quickly and accurately determine the order of nucleotides/bases of at least a part of the template sequence.

Specifically, in S11, the so-called bright spot set corresponding to the template includes multiple bright spots corresponding to the template, including the intensity and coordinate information of each bright spot.

The so-called coordinate mapping is to establish a mapping relationship between the original image, such as the set of bright spots corresponding to the template, and the target image, such as the image to be inspected. The mapping relationship here includes determining the coordinates of any bright spots in the original image in the mapped image. Location.

This embodiment places no restrictions on the method of determining coordinates or the method of implementing coordinate mapping. For coordinate mapping, for example, it can be achieved through Opencv's remap function. As for the determination of the coordinates of a bright spot, usually, a bright spot on an image occupies one or more pixels, and the coordinates of a certain pixel can be used as the coordinates of the bright spot, or the coordinates of the bright spot can be determined using, for example, a quadratic function interpolation method. The sub-pixel center coordinates of the bright spot are used as the coordinates of the bright spot.

Specifically, in some embodiments, the input image to be inspected may be a 16-bit tiff format image of 512*512 or 2048*2048, and the image in tiff format may be a grayscale image. For grayscale images, pixel values are the same as grayscale values. The input image can also be a color image. One pixel of the color image has three pixel values. The color image can be converted into a grayscale image, and then subsequent processing and detection can be performed to reduce the calculation amount and complexity of the image processing process. You can choose but are not limited to using floating point algorithm, integer method, shift method or average method to convert the non-grayscale image into a grayscale image.

The so-called bright spot collection corresponding to the template can be constructed when the base recognition is performed, or can be constructed and saved in advance. Here, a set of images collected from at least one round of sequencing is used to pre-construct a set of bright spots corresponding to the template and save them for later use.

In some examples, four nucleotides are labeled differently, and when sequenced, these different labels are excited to emit signals of different colors, and the different signals correspond to different types of nucleotides/bases. The so-called bright spot set corresponding to the template includes four bright spot sets corresponding to four nucleotides respectively.

In one example, a set of images from a round of sequencing is used to construct a set of bright spots corresponding to the template, including: sequentially or simultaneously adding four nucleotides to the reaction system for a round of sequencing to obtain the so-called A set of images, the set of images includes a first image, a second image, a third image and a fourth image, the first image, the second image, the third image and the fourth image are respectively collected from four kinds of nucleotides The signal emitted during the reaction, the so-called reaction system includes a template and a polymerase; perform bright spot detection on the first image, the second image, the third image and the fourth image respectively, and determine the bright spots in each image, including determining the bright spots coordinates; align the set of images so that the bright spots of the set of images are in the same coordinate system; merge the bright spots on the aligned set of images to obtain a first-level set of bright spots; according to the The first-level bright spot set is called a set of bright spots, and a set of bright spots corresponding to four nucleotides is established, that is, a template of four nucleotides/bases is established.

When constructing a bright spot set corresponding to the template, bright spots are detected and aligned on the set of images without order restrictions. The alignment of the set of images can be done by using the bright spots on the set of images, or without using the bright spots on the set of images. For example, some marks can be made at specific positions of the detection area, and the bright spots on the set of images can be aligned according to the Use some labeled information to align the set of images.

The so-called round of sequencing can include four base extensions, for example, four nucleotides are added in sequence to the reaction system to independently complete base extension including the collection of corresponding reaction signals, or it can include two base extensions, for example Four nucleotides are combined in pairs, and the nucleotides in each combination enter the reaction system at the same time for base extension. It can also include only one base extension. For example, four nucleotides are used in the reaction system for base extension at the same time. .

In one example, four nucleotides are added to the so-called reaction system at the same time, and the corresponding reaction signals are collected using an imaging system to obtain a set of images and/or images to be measured. The so-called imaging system includes a first laser , the second laser, the first camera and the second camera.

Further, the so-called template is DNA, and the four nucleotides carry a first label, a second label, a third label and a fourth label respectively, for example, four fluorescent molecules with different emission spectra or incomplete overlap; In a round of sequencing, a first laser is used to excite nucleotides, two of the four nucleotides respectively emit a first signal and a second signal, and the first camera and the second camera operate simultaneously to collect the first signal respectively. and a second signal, obtaining the first image and the second image, and using the second laser to excite the nucleotides, the other two nucleotides among the four nucleotides respectively emit the third signal and the fourth signal, the first The camera and the second camera operate synchronously to respectively collect the third signal and the fourth signal to obtain a third image and a fourth image. The so-called first laser and second laser can come from two lasers capable of emitting different wavelengths, or they can come from one laser capable of emitting multiple wavelengths.

Specifically, for example, the four deoxyribonucleotides dATP (sometimes abbreviated as A), dTTP (sometimes abbreviated as T), dGTP (sometimes abbreviated as G) and dCTP (sometimes abbreviated as C) respectively carry There are four fluorescent dyes: ATTO-532, ROX, CY5 and IF700. The spectral curves of these four fluorescent dyes are shown in Figure 2. The dotted curves from left to right are the absorption spectra of ATTO-532, ROX, CY5 and IF700 respectively. The peak wavelengths of each absorption spectrum are 531nm, 577nm, 651nm and 692nm respectively. The solid curves from left to right are the radiation spectra/emission spectra of ATTO-532, ROX, CY5 and IF700 respectively. The peak wavelengths of each radiation spectrum are 551nm, 602nm, 670nm and 712nm. When designing the optical path structure of the imaging system, taking into account the excitation efficiency of the dyes, lasers of at least two wavelengths are used to excite the four dyes in two time-sharing manners, and the two cameras pass through the dichroic mirror and dual bandpass The filter performs time-sharing fluorescence signal collection; in other words, the first laser and the second laser can operate asynchronously, and the first camera and the second camera can operate synchronously. In this way, the excitation of the four dyes and the corresponding signal can be efficiently achieved. collection.

The identification and detection of bright spots on the image is to detect signals from target molecules. This embodiment of the present invention does not limit the method of detecting bright spots. For example, the method disclosed in CN107918931A can be referred to.

In some embodiments, detecting bright spots includes using a k1*k2 matrix to detect each image in the set of images, including: determining the relationship between center intensity and edge intensity midS. A matrix that satisfies the first preset condition corresponds to a A so-called bright spot, the center intensity reflects the intensity of the central area of the matrix, the edge intensity reflects the intensity of the edge area of the matrix, a central area and an edge area form the so-called k1*k2 matrix, k1 and k2 are both greater than 1 is a natural number, and the k1*k2 matrix contains k1*k2 pixels.

The values of k1 and k2 are related to the density and distribution of template molecules on the solid matrix and the imaging resolution. It is generally expected that the k1*k2 matrix is not smaller than the size of a target bright spot. The so-called target bright spot corresponds to the target signal or corresponds to Target molecules/molecule clusters; preferably, it is generally expected that the k1*k2 matrix is smaller than the size of two independent bright spots on the image.

k1*k2 matrix, k1 and k2 can be equal or unequal. Generally, the value ranges of k1 and k2 are both greater than 1 and less than 10.

In one example, the relevant parameters of the imaging system are: the objective lens is 60 times, the size of the electronic sensor is 6.5 μm, the image formed by the microscope and then passes through the electronic sensor, the smallest size (resolution) that can be seen is about 0.1 μm, the obtained The image or input image can be a 16-bit grayscale or color image of 512*512, 1024*1024 or 2048*2048. A target bright spot corresponds to a single molecule, and the corresponding size is usually less than 10nm. The so-called single molecule includes a Or a few molecules/nucleic acid fragments, generally less than 10 molecules, such as 1, 2, 3, 4 or 5 molecules. A target bright spot occupies approximately 3*3 pixels on the image.

In another example, the relevant parameters of the imaging system are: the objective lens is 20 times, the image formed by the microscope then passes through the electronic sensor, the resolution is about 0.3μm, the image obtained or the input image can be 512*512, 1024*1024, For a grayscale or color image of 2048*2048 or 2560*2048, a target bright spot corresponds to a molecular cluster, and a target bright spot occupies approximately 5*5 pixels on the image.

k1 and k2 may be odd numbers or even numbers. In some embodiments, k1 and k2 are both odd numbers. In this way, it is convenient to set the central area and edge area of the matrix and facilitate subsequent calculations.

In one example, k1=k2=3.

The so-called central area and edge area are relative definitions. For example, an area of a certain size centered on the central pixel or central sub-pixel of the matrix can be the central area, and other areas constitute the edge area of the matrix.

The so-called intensity, or the intensity of the signal, including the center intensity and edge intensity here, is reflected in the image and is generally related to the size of the pixel, such as the pixel value of one or more pixels, the average or median value of multiple pixels. The number of bits, the sum of multiple pixel values, or is positively correlated with the pixel size.

In some examples, the so-called first preset condition is midS≥S1, midS=midInt-sumInts(1:n)/n, midInt represents the so-called central intensity, and sumInts(1:n)/n represents the so-called central strength. The edge strength of sumInts(1:n) represents the sum of pixel values from the 1st to Nth pixels in the edge area, n is a natural number not less than 4, and S1 is any value in [2,4]. The first preset condition was obtained by the inventor through training and summarizing a large amount of image data, and is suitable for bright spot detection in images with different signal intensities, bright spot densities and distributions from various sequencing platforms.

Specifically, k1 and k2 are both odd numbers greater than 3, and the so-called central area is a 3*3 area centered on the central pixel of the matrix. In an example, please refer to Figure 3, k1=k2=5, Figure 3 illustrates a 5*5 matrix, the central area is a 3*3 area centered on the pixel marked midS in the figure, and the central area is The pixel value of any pixel is the intensity of the central area (center intensity). For example, the pixel value of the pixel marked midS in the picture is the center intensity. n is 12, such as the pixels marked 1-12 in the picture. point, S1 takes 2. In this way, the bright spots corresponding to the target molecule can be quickly and effectively detected, which facilitates the construction of a bright spot collection corresponding to the template and facilitates the accurate identification of subsequent bases.

In other embodiments, bright spot detection includes: convolving each image in the set of images to obtain a convolved image; finding all the convolved images contained in the k3*k4 area For the peak pixel, k3 and k4 are both natural numbers greater than 1, and the k3*k4 area contains k3*k4 pixels of the convolved image; and, k5*k6 centered on the peak pixel is determined to meet the second preset condition. The area corresponds to a so-called bright spot. The second preset condition is that the pixel of the peak pixel in the k5*k6 area is not less than S2. Both k5 and k6 are natural numbers greater than 1. S2 can be performed through the pixels of the convolved image. Sure.

The image is convolved using a convolution kernel. The convolution kernel is also called a convolution template, filter, filter template or scan window. This embodiment does not limit the way to implement convolution. For example, after setting the convolution kernel, use Related functions in Matlab are performed. Convolving an image generally involves the calculation process of optionally flipping the convolution template, then sliding the convolution template on the original image, multiplying the elements at the corresponding positions and adding them up to obtain the final result. For example, so-called filtering can be implemented using Gaussian templates.

In some examples, the target molecules are nucleic acid molecule clusters, for example, nucleic acid molecule clusters formed after a nucleic acid molecule is amplified, such as strand displacement amplification or bridge amplification. The resolution of the imaging system used for image acquisition is about 0.3 μm. , set k3=k4=k5=k6=5; further, after studying the pattern and/or intensity changes of a large number of such target molecules on the image, the inventor set up a convolution kernel of 5*5 size To perform this convolution, the 5*5 size convolution kernel is shown in Figure 4. The mark on the convolution kernel shown in Figure 4 shows the coordinates/position of the pixel where the mark is located relative to the center pixel. The horizontal expression is x, vertically expressed as y, the unit is pixel, use such a 5*5 size convolution kernel to perform convolution operation on the image, including using the convolution kernel to reassign each pixel in the image. In this way, the difference between the center pixel and the edge pixel (for example, the outermost pixel) of the 5*5 area in the image can be enhanced.

Specifically, in one example, the inventor used a large amount of training data to set the intensity value/pixel value of the position/pixel without coordinate marking on the convolution kernel shown in Figure 4 to 0, and performed the operation through this convolution kernel. After the convolution operation set below, the intensity/pixel value Ints(x,y) of the pixel in the image, for example, the pixel with coordinates (x,y) becomes newInts(x,y), newInts(x,y)= (12*Ints(x,y)–Edge8Ints(x,y,2))*200/(Ints(x,y)+Edge8Ints(x,y,2)),

Ints(x,y) represents the pixel value/intensity value of the pixel/position with coordinates (x,y) before convolution; in order to facilitate fast operation, the range of newInts(x,y) can be further set to [0,255], newInts If (x, y) is less than 0, it is assigned a value of 0, and if it is greater than 255, it is assigned a value of 255;

Edge8Ints(x,y,2) represents the pixel values of 12 pixels in 8 directions (8 neighborhoods) of the center coordinate (x, y), and the distance (x, y) from the center coordinate is not less than 2 pixels. /Sum of intensity values, in this example, the so-called (x,y) distance from the center coordinate is not less than 12 pixels of 2 pixels, such as the pixel with coordinate mark shown in Figure 4, Edge8Ints(x ,y,2) can be expressed as Edge8Ints(x,y,2)=(Ints(x-2,y-1)+Ints(x-2,y)+Ints(x-2,y+1)+Ints (x+2,y-1)+Ints(x+2,y)+Ints(x+2,y+1)+Ints(x-1,y-2)+Ints(x,y-2)+ Ints(x+1,y-2)+Ints(x-1,y+2)+Ints(x,y+2)+Ints(x+1,y+2)), where Ints(x-2 ,y-1),Ints(x-2,y),Ints(x-2,y+1),Ints(x+2,y-1),Ints(x+2,y),Ints(x+ 2,y+1),Ints(x-1,y-2),Ints(x,y-2),Ints(x+1,y-2),Ints(x-1,y+2),Ints (x,y+2) and Ints(x+1,y+2) respectively represent the coordinates of (x-2,y-1), (x-2,y), (x-2,y+1), (x+2,y-1), (x+2,y), (x+2,y+1), (x-1,y-2), (x,y-2), (x+1 ,y-2), (x-1,y+2), (x,y+2) and (x+1,y+2) position/intensity value/pixel value before convolution.

Before performing this convolution, optionally perform Gaussian filtering on the image; then perform the above-mentioned convolution operation on the obtained Gaussian filtered image.

Figure 5 shows a comparison of images before and after convolution using the above method. The upper image is before convolution, and the lower image is after convolution. The boxes in the image indicate the images before and after convolution. Changes in signal intensity and/or pattern over the same area.

It can be understood that, according to needs, for example, if the morphology and/or intensity changes of the target molecules on the image have different characteristics, the size of the above-mentioned convolution kernel, the value in the convolution kernel, and the adjustment of, for example, Edge8Ints(x,y,n ), for adjusting the n, generally, if the size of the ideal bright spot is known to be m*m, it can be adjusted so that n=m/2 and rounded down.

For the settings of k3 and k4 or k5 and k6, similarly, the values of k3 and k4 or k5 and k6 are related to the density and distribution of template molecules on the solid matrix and the imaging resolution. It is generally expected that k3*k4 or k5* k6 is not less than the size of a target bright spot, which corresponds to the target signal or the target molecule/molecule cluster; preferably, it is generally expected that k3*k4 or k5*k6 is smaller than two independent bright spots on the image The size occupied.

k3 and k4, or k5 and k6 may or may not be equal. Generally, the value ranges of k3, k4, k5 and k6 are greater than 1 and less than 10.

In some embodiments, k3 is equal to k4, and/or, k5 is equal to k6.

In some embodiments, k3 and k4 are both odd numbers greater than 1, and/or k5 and k6 are both odd numbers greater than 1. Furthermore, for a target bright spot corresponding to a molecular cluster platform, for example, a template is amplified to form a molecular cluster, and the molecular cluster is fixed on the microsphere or the chip surface. Usually, the size of the molecular cluster is several Hundred nanometers, under the imaging light path of 20 times magnification, both k3 and k4 can be taken as odd numbers greater than 3, and/or both k5 and k6 can be taken as odd numbers greater than 3. In this way, calculation is facilitated, the construction of a bright spot set corresponding to the template is facilitated, and it is also conducive to the accurate identification of subsequent bases.

S2 is related to the pixels of the transformed image, for example, S2 can be determined by all pixels of the transformed image. In some embodiments, S2 is not less than the median of all pixels of the convolved image sorted in ascending order of pixel value, and/or is not greater than the eightieth percentile of all pixels of the transformed image sorted in ascending order of pixel value. number of digits. In one example, the input image is converted into a 256-color image (16-bit image), and S2 can be set to any value from 19 to 25. In this way, bright spots can be detected effectively.

In one example, perform Gaussian filtering on the original image and then perform the above convolution operation to obtain the convolved image; find all peak points (bright spots) on the convolved image, and ensure that the peak value is greater than a certain Value, for example, set a specific value to any value between 19 and 25. Generally, the larger the peak value, the brighter the point and the better the shape. Specifically, each pixel on the transformation map has a midS, bulge The value of MidS corresponding to the position is relatively high. Here, 25 is set as the filtering threshold. The position greater than 25 can be considered as a raised point; further, for all points that meet the above conditions, in the original The 3*3 area center of gravity method is used to determine its sub-pixel coordinates in the figure.

In some embodiments, after using the method of any of the above examples to detect bright spots on the image, the method further includes filtering the detected bright spots based on the intensity of the area where the bright spots are located on the original image. In this way, removing relatively dark or extremely bright spots or signals that are likely not or not simply from target molecules will help reduce the amount of calculations and increase the proportion of high-quality offline data.

In some embodiments, the so-called bright spot detection includes using a k7*k8 matrix to detect each image in the so-called set of images, including: determining that multiple pixels in a specified direction are monotonically fluctuating k7*k8 The matrix corresponds to a candidate bright spot; the candidate bright spot is screened using at least a part of the pixels in the corresponding k7*k8 matrix to determine the so-called bright spot. Both k7 and k8 are natural numbers greater than 1. The k7*k8 matrix Contains k7*k8 pixels.

Similarly, the values of k7 and k8 are generally related to the density and distribution of template molecules on the solid matrix and the imaging resolution. It is generally expected that k7*k8 is not less than the size of a target bright spot, and the so-called target bright spot corresponds to the target The signal may correspond to the target molecule/molecule cluster; preferably, it is generally expected that k7*k8 is smaller than the size occupied by two independent bright spots on the image.

In some examples, k7 is equal to k8, and/or both k7 and k8 are odd numbers greater than 1. In this way, calculation is facilitated, construction of a bright spot set corresponding to the template is facilitated, and subsequent base identification is facilitated.

The so-called specified direction can be any direction passing through the center of the k7*k8 matrix, such as the central pixel or sub-pixel; the so-called monotonic fluctuation means that the pixel values of multiple pixels in the specified direction have no fluctuation around the center of the k7*k8 matrix. , showing symmetrical fluctuations or almost symmetrical fluctuations.

In an example, please refer to Figure 6. Figure 6 shows that the pixel values of multiple pixels in the specified direction of a 5*5 matrix fluctuate monotonically. A specific specified direction can be the direction indicated by any arrow, a0, a1, a2 and a3 indicate the pixel values of the pixels where they are located, and this matrix corresponds to a candidate bright spot.

The so-called screening of the candidate bright spots by using at least part of the pixels in the corresponding k7*k8 matrix can further remove relatively dark or particularly bright spots, or remove the bright spots that are likely not or not simply from the target molecule. signal, which will help reduce the amount of calculation and increase the proportion of high-quality offline data.

For example, all pixels in the k7*k8 matrix corresponding to the candidate bright spot, the average value of the pixels in any row or column, or the high-frequency value are used as the background, and the intensity of the center of the candidate bright spot is compared with the size of the background, and the candidate To filter bright spots, for example, set the filtering condition to be that the intensity of the center of the candidate bright spot is not less than 3 times the background. Candidate bright spots that meet this condition are so-called bright spots. In this way, the proportion of high-quality data in the offline data can be increased.

In some examples, the bright spot detection further includes determining the sub-pixel coordinates of the detected bright spot using a center of gravity method. In this way, the coordinate information of the bright spot is obtained.

Systems that realize imaging based on optical path systems generally inevitably have chromatic aberration. Chromatic aberration generally causes a stationary signal to have different positions in multiple images collected at different time points; in addition, if the sequencing platform used is based on an imaging system and The relative movement of the chip requires multiple image acquisitions of a field of view on the chip. Image acquisition of the same field of view in different rounds of sequencing involves mechanical movement of related structures, which generally results in different images of the same field of view collected at different time points. Location. Aligning a set of images can at least correct positional deviations due to the above reasons at least to some extent.

In some embodiments, aligning the set of images includes using any image in the set of images, such as the coordinate system of the first image as a reference, to respectively align the second image, the third image and the fourth image. The coordinate system is transformed so that the coordinate system of the set of images is the same.

This embodiment of the present invention does not limit the method of converting the coordinate system. For example, MatLab correlation functions can be used.

Specifically, in one round of sequencing, four images of one field of view come from four bands of two cameras. Although optical adjustments have been made as much as possible, there is still a pixel shift (chromatic aberration) between the four images. Generally, the optical settings unchanged, it can be considered that the offset caused by the corresponding color difference is fixed; if this set of images comes from the first round of sequencing (cycle1) or previous rounds of sequencing, in cycle1 or previous rounds of sequencing, it generally corresponds to the four bases There is no crosstalk or the crosstalk is not obvious between the designated two of the four base signals. For example, ATGC carries four fluorescent dyes: ATTO-532, ROX, CY5 and IF700 respectively. In any of the previous rounds of sequencing, , at a certain time point, the first camera takes a picture of A and the second camera takes a picture of G at the same time. At another time point, the first camera takes a picture of T and the second camera takes a picture of C at the same time. The images collected from this round / signal, there is usually crosstalk between A and T signals or G and C signals, but there is no crosstalk or the crosstalk is not obvious between C and T or A and G signals. The so-called C and T signals do not have crosstalk or crosstalk. It is not obvious, which means that when a C signal is collected at a certain position, the T signal cannot be collected (C is bright and T is not bright). Therefore, in a certain sequencing, it is generally difficult to use one round of images in the first few rounds of the sequencing. Determine this fixed offset.

Thus, in some examples, a set of images is aligned, including using images from the Mth round of sequencing. M is greater than 20, 30 or 50, for example. One round of sequencing can generally determine the base type at a position on the template. When sequencing reaches cycle M, such as the 20th, 50th, 80th, 100th or 150th round, the emission spectra of fluorescent dyes partially overlap. Cross-color and/or phase imbalance caused by unsynchronized chemical reactions are generally more obvious due to accumulation or superposition, and are manifested as crosstalk between the signals of the four bases. The M-th round of collected signals can be used The image offset is determined to align the set of images.

In some examples, the images of the Mth round of sequencing include a fifth image, a sixth image, a seventh image and an eighth image. The so-called fifth image, sixth image, seventh image and eighth image are respectively related to the first image. The first image, the second image, the third image and the fourth image in the set of images correspond to the same kind of nucleotide.

In one example, the set of images that constructs the bright spot set corresponding to the template also comes from the Mth round, and the so-called fifth image, sixth image, seventh image and eighth image are respectively the same set of images. The first image, the second image, the third image and the fourth image in .

In one example, a set of images is said to be from cycle 1 and the offset is determined using images from cycle 100 of the same field of view to align the set of images. Specifically, for example, using the coordinate system of the fifth image as a reference, converting the coordinate systems of the sixth image, the seventh image, and the eighth image respectively may include: dividing the fifth image and the sixth image into one in the same manner. For blocks with a group size of k9*k10, k9 and k10 are both natural numbers greater than 30, and k9*k10 contains k9*k10 pixels; determine the offset of each block of the sixth image relative to the corresponding block of the fifth image. ;Based on this offset, align the second image with the first image. Similarly, the third image is aligned with the first image, and the fourth image is aligned with the first image to achieve alignment of the set of images quickly and accurately.

k9 and k10 may or may not be equal. The values of k9 and k10 are limited by the distribution, density and imaging resolution of target molecules/molecule clusters in the detection area. It is expected that the number of bright spots present on a k9*k10 block has statistical significance, for example, greater than 30, 50, 100 or 500.

Assuming that the offset of multiple images in a specific field of view caused by chromatic aberration is fixed, it can be understood that as long as there is signal crosstalk between each pair of images in a round of sequencing in that field of view, regardless of the nature of the crosstalk The shape of the signal on the image, the image of this round of sequencing can be used to determine the so-called fixed offset, and then align to construct a set of images corresponding to the bright spot set of the template. In some cases, grid signals regularly distributed horizontally and vertically can be set as feature information (source) in different detection areas, such as different chips or the same detection area. These feature information can be imaged in different channels or bands. That is, all base signals can be collected when collecting them, and the characteristic information including their distribution rules can be used to easily align the images. After dividing the image into multiple blocks, by aligning the feature information, the offset of a group of corresponding blocks can be determined.

Divide the image into blocks, and adjacent blocks may or may not overlap. In one example, adjacent blocks do not overlap and there is a common edge or vertex between adjacent blocks.

Figure 7 illustrates the process of dividing an image into multiple blocks and determining the offset between blocks to align a round of images in a field of view in one embodiment. The black square in the figure represents a so-called block. Specifically, Ground, using the coordinate system of a block on the image corresponding to base G (referred to as the G picture) as the base/reference, determine the offset of the corresponding block on the A picture, C picture and T picture relative to the block on the G picture. quantity.

During the test, it was found that the offsets between the corresponding blocks are not fixed, that is, the offsets of the two combined blocks located at different positions on the image are different. For example, the offsets of the two combined blocks located in the central areas of the two images are not the same. The offset of a block is 5 pixels (pixels), and the offset of two blocks located in the edge area of the two images is 10 pixels; moreover, the offset difference of adjacent block combinations is small. For example, for a 4112*2176 image, the offset of the long side 4112 is 4-5 pixels, and the offset of the short side 2176 is about 2-3 pixels. In an example, k9=k10=100. Generally, it can be considered that the offset is constant within a block of 100*100 size. Figure 8 shows that after dividing two images into blocks of 100*100 size, at least The offset between a part of the corresponding block combinations. The offset table shown in Figure 8 can represent the coordinate system relationship between the two images.

In some embodiments, the so-called merging of bright spots on a set of aligned images includes merging multiple bright spots within the preset range k11*k12 into one bright spot, where both k11 and k12 are greater than 1. Natural numbers, k11*k12 contains k11*k12 pixels.

Generally, k11*k12 is set to be no larger than the size of two separate target bright spots. Preferably, k11*k12 is set to be no larger than the size of one target bright spot.

In an imaging system, for example, the size of the electronic sensor is 6.5 μm, the microscope magnification is 60 times, and the resolution is 0.1 μm. The size of a bright spot corresponding to the target molecule including the molecular cluster is generally less than 10*10 or 5* 5. You can set k11=k12=3, that is, set the so-called preset range to 3*3 to perform bright spot merging. In this way, a bright spot set corresponding to the template can be accurately constructed.

Specifically, when merging bright spots within a preset range, you can first set a blank set/blank image/blank template (TemplateVec), and then sequentially combine the first image, the second image, the third image or the fourth image ( The bright spots on the picture (referred to as picture A, picture C, picture G and picture T) are marked on the blank picture. When marking a bright spot, if it is found that there is a bright spot nearby (within the preset range) , the weight of the two bright spots can be used to determine the position of the new bright spot after merging the two bright spots. For example, the intensity of bright spot 1 is 350, and the coordinates are (3.0,5.0), and bright spot 2 The intensity is 150 and the coordinates are (4.0,7.0). Mark these two bright spots as a new bright spot. The intensity of the new bright spot is 290 and the coordinates are (3.3,5.6). In this way, the bright spots that meet the preset conditions on the set of images are combined to facilitate obtaining a set of bright spots corresponding to the template.

Please refer to Figure 9. Figure 9 illustrates a process of constructing a set of bright spots corresponding to a template using a set of images from a round of sequencing in one embodiment, including the process of constructing a set of images A, C, and G in a set of images. The bright spots on the picture and T picture are detected and identified to obtain the bright spot set of each image. Using the coordinate system of the G picture as the reference coordinate system, aligning the set of images includes merging the bright spot sets of each picture to obtain a first-level bright spot. Spot set, and then through coordinate system conversion, convert the coordinate system of the first-level bright spot set into the original coordinate system of the A picture, C picture, G picture and T picture, and obtain the bright spots corresponding to the four nucleotides/bases Set, that is, templates of four nucleotides/bases are obtained.

In this embodiment, the intensity in S21 is the intensity after correction. In some embodiments, the correction intensity includes cross-color correction and/or phase correction.

Specifically, before correcting the intensity of the corresponding coordinate position on the image to be inspected, the image to be inspected is aligned with the so-called bright spot set corresponding to the template. This will facilitate subsequent steps.

In one example, ATGC contains four fluorescent dyes: ATTO-532, ROX, CY5 and IF700. During sequencing, lasers of two bands are used to excite the four fluorescent dyes respectively. After each excitation, two cameras are used to simultaneously Collect fluorescence signals; Figure 10 shows the crosstalk diagram between four images of one field of view in the 50th round of sequencing based on this example. From top to bottom and from left to right, the crosstalk scatter diagram of bases A-C ( The abscissa is the relative intensity of the A signal, the ordinate is the relative intensity of the C signal), base A-G crosstalk scatter plot (the abscissa is the relative intensity of the A signal, the ordinate is the relative intensity of the G signal), base A- Tcrosstalk scatter plot (the abscissa is the relative intensity of the A signal, the ordinate is the relative intensity of the T signal), base C-G crosstalk scatter plot (the abscissa is the relative intensity of the C signal, the ordinate is the relative intensity of the G signal) , base C-T crosstalk scatter plot (the abscissa is the relative intensity of the C signal, the ordinate is the relative intensity of the T signal) and the base G-T crosstalk scatter plot (the abscissa is the relative intensity of the G signal, the ordinate is the T The relative strength of the signal), a point on each scatter plot represents a position of the corresponding coordinates on the so-called image to be inspected; it can be seen from the two arms of each crosstalk scatter plot and the dispersion of the points on the plot, The A signal (A picture) is more obviously affected by the crosstalk of the T signal, and the C signal (C picture) is more obviously affected by the G signal. This is manifested in the fact that the positions of the corresponding coordinates on multiple A pictures have obvious T signals, and multiple C There are obvious G signals at the corresponding coordinates on the map.

In some examples, correction of intensity includes crosstalk correction based on at least one of the images from the same round of sequencing, the same field of view, and corresponding to different kinds of nucleotides/bases.

Correct crosstalk to facilitate accurate identification of bases. In some examples, the image Xi and the image to be inspected come from the same round of sequencing, the image Xi and the image to be inspected correspond to the same field of view, the image to be inspected is subject to crosstalk by the signal of the nucleotide corresponding to the image Xi, and the image to be inspected is cross-colored. Correction includes: fitting signals at positions of multiple corresponding coordinates in a specific area of the image to be inspected to obtain a fitting result; and correcting signals at positions of corresponding coordinates on the image to be inspected based on the fitting results. In this way, the signal crosstalk from the bases corresponding to the image Xi on the image to be detected can be eliminated, so that the signals in the image to be detected correspond to only one kind of base as much as possible, which is conducive to the accurate identification of bases and the accurate determination of nuclei. The sequence of nucleotides.

If there are no exceptions, "AC correction" or "A->C" or "A-C" represents the crosstalk of the A signal at the corresponding coordinate position of the corrected C chart (that is, the crosstalk of the corrected A signal to the C signal); similarly , "TA correction" or "T->A" means correcting the crosstalk of the T signal at the corresponding coordinate position on the A map (that is, correcting the crosstalk of the T signal to the A signal), "CG correction" or "C->G ” means correcting the crosstalk of the C signal at the corresponding coordinate position on the G map (that is, correcting the crosstalk of the C signal to the G signal) and so on.

Four-dimensional data are corrected in pairs. There are 12 situations. The correction process can be expressed as of which/> is the crosstalk correction coefficient matrix. The value in the crosstalk matrix represents the fitting result (correction coefficient) of the pair of signals. For example, R _AC represents the fitting result based on the correction of the crosstalk of the A signal at the corresponding coordinate position on the C diagram. /Correction coefficient, is the observed value,/> is the true value (corrected value).

The so-called specific area may be the entire image to be inspected or a part of the image to be inspected. Preferably, the specific area is selected from at least a part of the central area of the image to be inspected. The central area of the image can be generally understood. For example, for an image with a size of 4000*2000, the central area of the image can be 3000*1500. , 2056*1024, 2000*1500, 1024*1024, 1024*512, 1000*500, 1000*1000, 512*512 or 512*256, etc. In contrast, other areas of the image can be called edge areas. Generally, the intensity value of the corresponding coordinate position in the central area of the image fluctuates less and appears to be relatively convergent on the crosstalk diagram, such as the points in the black circles in the A-G crosstalk scatter diagram in Figure 11. Using the fitting results/correction coefficients determined by fitting the intensity values of at least a part of the region for correction, chromatic aberration correction can be achieved quickly and accurately.

This embodiment does not limit the fitting method. For example, MatLab cftool curve fitting toolbox, aTool, CurveExpert and other software can be used; the fitting can be linear fitting or nonlinear fitting. There are no special restrictions on the amount of data or sampling amount used for fitting, that is, how many intensity positions of corresponding coordinates in a specific area on the image are selected to do the fitting. In principle, as long as the desired fitting can be solved Y coefficients of the Y-element equation are enough. For example, for linear fitting, you can take 2, 5, 10, 20, 30 or 50 coefficients; preferably, you hope that the sampling amount can be statistically significant, for example No less than 20, 30 or 50; optionally, in order to prevent the calculation amount from being too large, the sampling amount can be limited to less than 200 or less than 100. In this way, correction can be accurately achieved using the corresponding fitting results (correction coefficients).

In some examples, a linear fit is performed. In this way, it is easy to calculate, takes less time, and is conducive to rapid correction.

Specifically, please refer to Figure 12 and Figure 13. In one example, the image to be inspected is picture A, and the image Xi is picture T. The intensity of the signals at the 20 corresponding coordinate positions of the central area on the image to be inspected is selected for linear calculation. Fitting. Figure 12 shows the results of the fitting. The abscissa is the relative signal intensity value of A, and the ordinate is the relative signal intensity value of T. The fitting result determines the slope k of the fitted straight line. With this The slope is used as a correction coefficient to correct the signal intensity at each corresponding coordinate position of the image to be examined. For example, I _T' = I _T - _{I A} × k, I _T' is the T signal intensity at the position after correction, I _T is the observed T signal intensity (observed value) at this position, I _A is the observed A signal intensity (observed value) at this position; Figure 13 illustrates the positions of the 20 corresponding coordinates on the image to be inspected The signal before correction and the result after correction using this method. In this way, the contribution of the T signal to the fluctuation of the signal intensity at the position of the corresponding coordinates of the image to be inspected can be eliminated or reduced, and a corrected image to be inspected can be obtained.

Compared with Figure 10, Figure 14 shows the crosstalk diagram between four images of the same field of view for this round of sequencing after chromatic aberration correction using the above example. It can be seen that after the chromatic aberration correction, the signal crosstalk between the images corresponding to different bases in the same field of view and in the same round is significantly reduced, which is beneficial to the accurate identification of bases and the detection of longer sequences.

Please refer to Figure 15-18. Figure 15-18 shows an example of the signal crosstalk between two images of adjacent rounds corresponding to the same base in the same field of view. A point on the graph represents a so-called corresponding coordinate. The position of The signal strength relationship diagram of G and two T pictures. The four phasing scatter diagrams in Figure 16 are the signal intensity relationships of two A pictures, two C pictures, two G and two T pictures of cycle30 and cycle31 respectively. Figure, the four phasing scatter diagrams in Figure 17 are the signal intensity relationship diagrams of two A pictures, two C pictures, two G pictures and two T pictures of cycle60 and cycle61 respectively. The four phasing scatter diagrams in Figure 18 The dot plots are the signal strength relationship diagrams of two A pictures, two C pictures, two G pictures and two T pictures of cycle90 and cycle91 respectively.

It can be seen that in this example, the phasing or prephasing of C or T is more obvious relative to A or G; and as the number of sequencing rounds increases, the phase imbalance of the chemical reactions of various bases is caused. The signal crosstalk is becoming more and more serious. It can be seen from Figure 18 that by the 91st round of sequencing in this example, the phase loss has made it difficult to accurately distinguish whether the signal at a certain position in the T chart comes from the 90th round of sequencing or the 91st round. Sequencing. Generally, at the end of the sequencing process, all the corresponding coordinate positions will be bright and uniform. In this case, the correct base cannot be identified, which means that sequencing cannot continue. Dephasing is the limiting edge. The main reason for read length in side-of-synthesis sequencing.

The upper and lower graphs in Figure 19 respectively show the relationship between the phasing ratio or prephasing ratio of the four bases in a certain nucleic acid sample sequencing and the number of sequencing rounds. As the number of sequencing rounds increases, the phasing and prephasing ratios of each base increase. The proportion of prephasing increases.

Perform phasing or prephasing correction to facilitate the correct identification of bases and read longer sequences. Phase correction can be performed before crosstalk correction or after crosstalk correction.

In some examples, the correction of the intensity includes a phase correction based on at least one of the images from adjacent rounds of sequencing and corresponding to the same kind of nucleotide.

Specifically, in one example, the image Yj and the image to be inspected come from two adjacent rounds of sequencing. For example, the image Yj comes from the 31st round of sequencing, and the image to be inspected comes from the 30th round of sequencing. Yj and the image to be inspected correspond to the same field of view, The image Yj and the image to be tested correspond to the same type of nucleotide/base, such as A. The so-called phase correction includes: fitting the signals at the positions of multiple corresponding coordinates in a specific area of the image to be tested to obtain the fitting result. ; and a signal that corrects the position of the corresponding coordinates on the image to be inspected based on the fitting results.

Similarly, the specific area referred to here may be the entire image to be inspected or a part of the image to be inspected. Preferably, the specific area is selected from at least a part of the central area of the image to be inspected. The central area of the image can be generally understood. For example, for an image with a size of 4000*2000, the central area of the image can be 3000*1500. , 2056*1024, 2000*1500, 1024*1024, 1024*512, 1000*500, 1000*1000, 512*512 or 512*256, etc. In contrast, other areas of the image can be called edge areas. Generally, the intensity value of the corresponding coordinate position in the central area of the image fluctuates less and appears more convergent on the phasing scatter plot, as shown in the phasing scatter plot of cycle 30 and cycle 31 in Figure 20. Dots in black circles. Phase correction can be achieved quickly and accurately by using the fitting result/correction coefficient determined by fitting the intensity values of at least a part of the region.

Similarly, this embodiment does not limit the fitting method; the fitting may be linear fitting or nonlinear fitting. There are no special restrictions on the amount of data or sampling amount used for fitting, that is, how many intensity positions of corresponding coordinates in a specific area on the image are selected to do the fitting. In principle, as long as the desired fitting can be solved Y coefficients of the Y-element equation are enough. For example, for linear fitting, you can take 2, 5, 10, 20, 30 or 50 coefficients; preferably, you hope that the sampling amount can be statistically significant, for example No less than 20, 30 or 50; optionally, in order to prevent the calculation amount from being too large, the sampling amount can be limited to less than 200 or less than 100. In this way, correction can be accurately achieved using the corresponding fitting results (correction coefficients).

In some examples, linear fitting is performed to correct phasing before crosstalk correction according to the method of the above example, and the R^2 of the linear fitting is 0.97; in other examples, signals at the same multiple positions are taken for fitting, According to the method of the above example, linear fitting is performed after crosstalk correction to correct phasing. The R^2 of the linear fitting is 0.93.

For S31, in some examples, the so-called intensity of the signal at the corresponding coordinate position on the image to be detected is an array (four-dimensional data) containing four values, corresponding to the four nucleotides/bases at the position. The signal intensity, for example, can be expressed as {Ints A, Ints T, Ints G, Ints C}, Ints A, Ints T, Ints G and Ints C represent the signal intensity values of bases A, T, G and C respectively, after correction Finally, generally, Ints A, Ints T, Ints G and Ints C have the same basis. The maximum value (max) in the array can be compared with the so-called first preset value, which is greater than or equal to the first preset value. By setting the value, it can be determined that the base type corresponding to the position on the image is the base corresponding to the maximum value, that is, it is identified that the base at the corresponding position on the corresponding nucleic acid molecule is the base corresponding to the maximum value. base; if the maximum value (max) in the array is less than the first preset value, it can be determined that the base type corresponding to the position on the image cannot be accurately identified, and the base type at the position of the corresponding nucleic acid molecule can be The base is recorded as N or a gap, and N is any one of ATGC; in some examples, the reads containing N or gaps after base identification can be further processed, for example, based on other reads For example, the information of adjacent reads can be used to infer the base type represented by the N or gap in the read, or be partially filtered out, etc., to improve the utilization rate of the output data or the quality of the output data.

In some examples, each value in {Ints A, Ints T, Ints G, Ints C} is a value after processing, such as normalization.

In one example, the quality score (QualityScore, QScore for short) is calculated on the four-dimensional data. QScore is essentially a prior probability and can be calculated using known methods. For example, refer to [Ewing et al., Base-calling of automated sequencer traces using phred.I.Accuracy assessment.,Genome Res.1998Mar,8(3):175-85.]. Here, the inventor uses the ratio of the maximum value in the corrected 4-dimensional data to the total value to calculate the QScore. The range of the calculated QScore size is [0,40]. Specifically, QScore=(1.0*maxInts/sumInts -0.25)/0.75*40, maxInts is the maximum value among Ints A, Ints T, Ints G and Ints C, sumInts is the sum of Ints A, Ints T, Ints G and Ints C. Correspondingly, set the first preset The value is 0.1. If the QScore is greater than 0.1, it is determined that the base type at this position is the base corresponding to maxInts. In this way, base recognition can be effectively performed.

The above logic and/or steps represented in the flowcharts or otherwise described herein may be considered as a sequence listing of executable instructions for implementing logical functions, and may be embodied in any computer-readable storage medium to For use by, or in conjunction with, instruction execution systems, devices or devices (such as computer-based systems, systems including processors or other systems that can fetch instructions from and execute instructions). equipment for use. For example, a computer-readable storage medium in an embodiment of the present invention is used to store a program for computer execution, and executing the program includes completing the method of any of the above embodiments. The computer-readable storage medium may be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, device, or device, including but not limited to read-only storage media. memory, magnetic disk or optical disk, etc. More specifically, the computer readable storage medium includes the following (non-exhaustive list): electrical connection with one or more wires (electronic device), portable computer disk case (magnetic device), random access memory (RAM) , read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable compact disc read-only memory (CDROM). Additionally, the computer-readable storage medium may even be paper or other suitable media on which the program may be printed, for example by optical scanning of the paper or other medium, followed by editing, interpretation, or other suitable means if necessary Processing is performed to obtain a program electronically and then stored in computer memory. The above description of the technical features and advantages of the base identification method in any embodiment is also applicable to the computer-readable storage medium, and will not be described again here.

Further, an embodiment of the present invention provides a computer product, including the computer-readable storage medium in any of the above embodiments.

For example, an embodiment of the present invention provides a system, including the computer product provided in any of the above embodiments and at least one processor, for executing a program stored in the computer-readable storage medium.

For example, embodiments of the present invention provide a computer program product, including instructions for identifying one or more bases on a nucleic acid. When the computer executes the program, the instructions cause the computer to perform the identification of nucleic acids in any of the above embodiments. method of one or more bases.

Embodiments of the present invention provide a system configured to perform the method of identifying one or more bases in a nucleic acid in any of the above embodiments.

Referring to Figure 21, an embodiment of the present invention provides a system 100, which includes a plurality of modules and is used to perform the steps of the method for identifying one or more bases in a nucleic acid in any of the above embodiments. The system 100 includes a mapping module 110 , a signal determination module 120 and a comparison module 130 . The mapping module 110 is used to map the coordinates of each bright spot in the bright spot set corresponding to the template onto the image to be inspected, and determine the position of the corresponding coordinates on the image to be inspected. The so-called bright spot set corresponding to the template is constructed based on a set of images. Each image in the set of images contains multiple bright spots; the set of images and the image to be detected are both derived from sequencing and correspond to An identical field of view; sequencing includes adding nucleotides for multiple rounds of sequencing, the set of images is from at least one round of sequencing, and at least a portion of the signal appears as at least a portion of the bright spots on the set of images ;

The signal determination module 120 is used to determine the intensity of the signal at the position of the corresponding coordinate on the image to be detected from the mapping module 110. The so-called intensity is the corrected intensity; and the comparison module 130 is used to compare the signal from the signal determination module 120. The intensity of the signal at the position of the corresponding coordinate on the image to be detected is compared with the size of the first preset value, and the base type corresponding to the position is determined based on the comparison result to realize the base identification.

Those skilled in the art know that in addition to implementing the controller/processor in the form of pure computer-readable program code, the controller can be controlled by logic gates, switches, application-specific integrated circuits, or editable logic by converting the method steps into logic. The same function can be achieved in the form of processors and embedded microcontrollers. Therefore, this controller/processor can be considered as a hardware component, and the devices included therein for implementing various functions can also be considered as structures within the hardware component. Or even, the means for implementing various functions can be regarded as either software modules that implement the methods or structures within hardware components.

The description of the technical features and advantages of the method for identifying one or more bases in nucleic acids in the above embodiments is also applicable to this system and will not be described again here. It can be understood that additional technical features of the method for identifying one or more bases in nucleic acids in any of the above embodiments, including sub-steps, additional steps, alternative or better settings or processing, etc., can be achieved by making The system or modules of the system further comprise units/modules or sub-units/sub-modules for implementation.

In some examples, the system 100 further includes a bright spot set building module 140 for building a so-called bright spot set corresponding to the template, and the bright spot set building module 140 is connected to the mapping module 110 .

In other examples, the mapping module 110 includes a bright spot set construction sub-module 111 for constructing a bright spot set corresponding to the template. The bright spot set construction sub-module 111 includes: an image acquisition unit 1111 for sequentially or simultaneously adding four Add nucleotides to the reaction system for a round of sequencing to obtain a set of images; the four nucleotides carry different labels and are excited to emit signals of different colors. The set of images includes the first image, the second image, the third image and the fourth image, the first image, the second image, the third image and the fourth image are respectively collected from the reaction signals of four nucleotides in the same field of view, and the reaction system includes the template and the polymerase; The bright spot detection unit 1113 is used to detect bright spots on the first image, the second image, the third image and the fourth image from the image acquisition unit 1111 respectively, and determine the bright spots of each image; the alignment unit 1115 is used to Align the so-called set of images; the merging unit 1117 is used to merge the bright spots on the aligned set of images from the alignment unit 1115 to obtain a first-level bright spot set; and the bright spot set creation unit 1119 is used to Based on the first-level bright spot set from the merging unit 1117, bright spot sets respectively corresponding to the four nucleotides are established.

In some examples, the image acquisition unit 1111 adds four nucleotides to the reaction system at the same time and uses an imaging system to collect the signals to obtain the set of images and/or the image to be tested, so-called The imaging system includes a first laser, a second laser, a first camera and a second camera.

In some examples, the four nucleotides added by the image acquisition unit 1111 carry a first label, a second label, a third label and a fourth label respectively. In a so-called round of sequencing, the first laser excitation is turned on. Described nucleotides, two of the four nucleotides respectively emit a first signal and a second signal, the first camera and the second camera operate synchronously to respectively collect the first signal and the second signal, and obtain the first signal and the second signal respectively. A first image and a second image, and, turning on the second laser to excite the nucleotides, the other two nucleotides among the four nucleotides respectively emit a third signal and a fourth signal, the first camera and the second camera Synchronize operations to separately collect the third signal and the fourth signal to obtain a third image and a fourth image.

In some examples, the bright spot detection unit 1113 uses the k1*k2 matrix to detect each image in the set of images, including determining the relationship between the center intensity and the edge intensity midS. The matrix that satisfies the first preset condition corresponds to a For the bright spot, the center intensity reflects the intensity of the center area of the matrix, and the edge intensity reflects the intensity of the edge area of the matrix. The center area and the edge area constitute the matrix, and k1 and k2 are both natural numbers greater than 1. , the k1*k2 matrix contains k1*k2 pixels.

In some examples, when the bright spot detection unit 1113 uses a k1*k2 matrix to detect each image in the set of images, k1 is equal to k2.

In some examples, when the bright spot detection unit 1113 uses the k1*k2 matrix to detect each image in the set of images, both k1 and k2 are odd numbers greater than 1.

When the bright spot detection unit 1113 uses the k1*k2 matrix to detect each image in the set of images, k1 and k2 are both odd numbers greater than 3, and the central area is centered on the central pixel of the matrix. 3*3 area.

In some examples, when the bright spot detection unit 1112 uses the k1*k2 matrix to detect each image in the set of images, the first preset condition is midS≥S1, midS=midInt-sumInts(1:n) /n, midInt represents the center intensity, sumInts(1:n)/n represents the edge intensity, sumInts(1:n) represents the sum of pixel values of the 1st to Nth pixels in the edge area, n is a natural number not less than 4, and S1 is any value in [2,4].

In some examples, the bright spot detection unit 1113 includes: convolving each image in the set of images to obtain a convolved image; finding all the convolved images in k3 The *k4 area contains peak pixels. Both k3 and k4 are natural numbers greater than 1. The k3*k4 area contains k3*k4 pixels of the convolved image; determine the pixels centered on the peak pixel that meet the second preset condition. The k5*k6 area corresponds to one of the bright spots. The second preset condition is that the peak pixel of the k5*k6 area is not less than S2. Both k5 and k6 are natural numbers greater than 1. S2 can pass the convolution. The pixels of the resulting image are determined.

In some examples, k3 is equal to k4, and/or k5 is equal to k6.

In some examples, k3 and k4 are both odd numbers greater than 1, and/or k5 and k6 are both odd numbers greater than 1.

In some examples, k3 and k4 are both odd numbers greater than 3, and/or k5 and k6 are both odd numbers greater than 3.

In some examples, S2 is not less than the median of all pixels of the convolved image sorted in ascending order of pixel value, and/or is not greater than the median of all pixels of the convolved image sorted in ascending order of pixel value. Eighty percentile.

In some examples, the bright spot detection unit 1113 further includes filtering the bright spots of the image based on the intensity of the area where the bright spot is located on the original image.

In some examples, the bright spot detection unit 1113 includes using a k7*k8 matrix to detect each image in the set of images, including determining that multiple pixel values in the specified direction are monotonically fluctuating. The k7*k8 matrix corresponds to a Candidate bright spots are screened using pixels in at least part of the area in the corresponding k7*k8 matrix to determine the bright spots, k7 and k8 are both natural numbers greater than 1, and the k7*k8 matrix contains k7* k8 pixels.

In some examples, k7 is equal to k8, and/or both k7 and k8 are odd numbers greater than 1.

In some examples, the mapping module 110 also includes a sub-pixel coordinate confirmation sub-module 112 for determining the sub-pixel coordinates of the bright spot using the center of gravity method.

In some examples, alignment unit 1115 performs this alignment using images from the Mth round of sequencing.

In some examples, the so-called set of images is aligned, including using images from the Mth round of sequencing. M is greater than 20, 30 or 50, for example.

In some examples, when the alignment unit 1115 performs the alignment using images from the M-th round of sequencing, the images from the M-th round of sequencing include fifth images, sixth images, seventh images, and eighth images, the so-called fifth images. , the sixth image, the seventh image and the eighth image respectively correspond to the reaction signals of the same nucleotides as the first image, the second image, the third image and the fourth image. The coordinate system of the so-called fifth image is The benchmark, respectively converting the coordinate systems of the sixth image, the seventh image and the eighth image, includes dividing the fifth image and the sixth image into a group of blocks of size k9*k10 in the same way, k9 and k10 are all natural numbers greater than 30, and k9*k10 includes k9*k10 pixels; the offset of each block of the so-called sixth image relative to the corresponding block of the so-called fifth image is determined respectively; based on the offset The shift amount aligns the second image with the first image.

In some examples, the merging unit 1117, when merging the bright spots on the aligned set of images, includes merging multiple bright spots within the preset range k11*k12 into one bright spot, where both k11 and k12 are greater than A natural number of 1, k11*k12 contains k11*k12 pixels.

In some examples, the signal determination module 120 is used to determine the intensity of the signal at the position of the corresponding coordinate on the image to be inspected. The intensity is the corrected intensity, and the corrected intensity includes cross-color correction and/or phase correction.

In some examples, before the signal determination module 120 corrects the intensity, the mapping module 110 aligns the image to be detected with the set of bright spots corresponding to the template.

In some examples, when correcting the intensity, the signal determination module 120 includes using cross-color correction based on at least one of the images from the same round of sequencing and corresponding to different types of nucleotides.

In some examples, when the signal determination module 120 performs intensity correction, the cross-color correction used includes:

Fit the signals of the positions of multiple corresponding coordinates in the specific area of the image to be inspected to obtain a fitting result; and correct the signals of the positions of the corresponding coordinates on the image to be inspected based on the fitting results. The image Xi and the image to be tested are from the same round of sequencing, the image Xi and the image to be tested correspond to the same field of view, and the image to be tested contains signals from the nucleotides corresponding to the image Xi.

In some examples, the fit is a linear fit.

In some examples, when correcting the intensity, the signal determination module 120 includes using phase correction based on at least one of the images from adjacent rounds of sequencing and corresponding to the same type of nucleotide.

In some examples, when the signal determination module 120 performs intensity correction, the phase correction used includes: fitting signals at the positions of multiple corresponding coordinates in a specific area of the image to be inspected to obtain a fitting result; and correcting the signal of the position of the corresponding coordinate on the image to be inspected based on the fitting relationship. The image Yj and the image to be tested are from two adjacent rounds of sequencing, the image Yj and the image to be tested correspond to the same field of view, and the image Yj and the image to be tested correspond to the same type of nucleotides.

Using the method, product and/or system of any embodiment of the present invention to perform base identification can quickly and accurately identify bases and achieve the determination of the nucleotide/base sequence of at least a part of the template sequence.

In the description of this specification, reference to the description of the terms "one embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples" is intended to be in conjunction with the description. An embodiment or example describes a specific feature, structure, material, or characteristic that is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

Although the embodiments of the present invention have been shown and described above, it can be understood that the above-mentioned embodiments are illustrative and should not be construed as limitations of the present invention. Those of ordinary skill in the art can make modifications to the above-mentioned embodiments within the scope of the present invention. The embodiments are subject to changes, modifications, substitutions and variations.

Claims (51)

1. A method of identifying one or more bases in a nucleic acid by detecting an image obtained from sequencing, comprising:

mapping coordinates of each bright spot in a bright spot set corresponding to a template onto an image to be detected, wherein the mapping comprises determining the coordinate position of any bright spot in the bright spot set corresponding to the template on the image to be detected so as to determine the position of corresponding coordinates on the image to be detected;

determining the intensity of a signal of the position of the corresponding coordinate on the image to be detected, wherein the intensity is corrected intensity; and

comparing the intensity of the signal of the position of the corresponding coordinate on the image to be detected with a first preset value, and judging the base type corresponding to the position based on the comparison result to realize the base identification;

the set of bright spots corresponding to the template is constructed based on a set of images, each image of the set of images comprising a plurality of bright spots,

The set of images and the image to be examined are both from a sequence that includes the addition of nucleotides and corresponds to one and the same field of view,

the sequencing includes multiple rounds of sequencing, the set of images from at least one round of sequencing, at least a portion of the signals appearing as at least a portion of the bright spots on the set of images.

2. The method of claim 1, wherein in performing said sequencing, four nucleotides are added, the four nucleotides bearing different labels, said different labels being excited to emit signals of different colors, and constructing said set of hot spots corresponding to templates comprises:

sequentially or simultaneously adding four nucleotides into a reaction system to perform one-round sequencing to obtain a group of images, wherein the group of images comprises a first image, a second image, a third image and a fourth image, the first image, the second image, the third image and the fourth image are respectively acquired from reaction signals of the four nucleotides in the same visual field, and the reaction system comprises the template and polymerase;

respectively carrying out bright spot detection on the first image, the second image, the third image and the fourth image to determine bright spots of the images;

Aligning the set of images;

combining the bright spots on the aligned group of images to obtain a first-level bright spot set;

and establishing a set of bright spots respectively corresponding to the four nucleotides according to the first-level bright spot set.

3. The method of claim 2, wherein four nucleotides are added simultaneously to the reaction system, the signals being acquired with an imaging system comprising a first laser, a second laser, a first camera and a second camera to obtain the set of images and/or the image to be examined.

4. The method of claim 3, wherein the template is DNA and the four nucleotides bear a first tag, a second tag, a third tag, and a fourth tag, respectively, and wherein, during the one round of sequencing,

starting a first laser to excite the nucleotide, respectively emitting a first signal and a second signal by two of the four nucleotides, synchronously operating the first camera and the second camera to respectively acquire the first signal and the second signal, obtaining a first image and a second image,

and starting a second laser to excite the nucleotides, wherein the other two nucleotides in the four nucleotides respectively emit a third signal and a fourth signal, and the first camera and the second camera synchronously work to respectively acquire the third signal and the fourth signal, so as to obtain a third image and a fourth image.

5. The method of any of claims 2-4, wherein the speckle detection comprises detecting each image in the set of images using a k1 x k2 matrix, comprising,

the method comprises the steps that a matrix, in which the relation midS between center intensity and edge intensity meets a first preset condition, corresponds to one bright spot, the center intensity reflects the intensity of a center area of the matrix, the edge intensity reflects the intensity of an edge area of the matrix, the center area and the edge area form the matrix, k1 and k2 are natural numbers larger than 1, and a k1 k2 matrix comprises k1 k2 pixels.

6. The method of claim 5, wherein k1 is equal to k2.

7. The method of claim 5, wherein k1 and k2 are each an odd number greater than 1.

8. The method of claim 7, wherein k1 and k2 are each an odd number greater than 3, and the center region is a 3*3 region centered on a center pixel of the matrix.

9. The method of claim 5, wherein the first predetermined condition is midS.gtoreq.S 1,

mids=midint-sumInts (1:n)/N, midInt represents the center intensity, sumInts (1:n)/N represents the edge intensity, sumInts (1:n) represents the sum of pixel values of 1 st to nth pixels of the edge region, N is a natural number not less than 4, and S1 is any value of [2, 4 ].

10. The method of any of claims 2-4, wherein the hot spot detection comprises:

respectively convolving each image in the group of images to obtain convolved images;

searching all pixels containing peaks in a k3 k4 region in the convolved image, wherein k3 and k4 are natural numbers larger than 1, and the k3 k4 region contains k3 k4 pixels of the convolved image;

and determining that a k5 x k6 region with a peak pixel as a center corresponds to one bright spot, wherein the second preset condition is that the pixel of the peak pixel of the k5 x k6 region is not less than S2, k5 and k6 are natural numbers which are larger than 1, and S2 can be determined through the pixel of the convolved image.

11. The method of claim 10, wherein k3 is equal to k4 and/or k5 is equal to k6.

12. The method according to claim 10, wherein k3 and k4 are each odd numbers greater than 1, and/or k5 and k6 are each odd numbers greater than 1.

13. The method of claim 12, wherein k3 and k4 are each an odd number greater than 3, and/or k5 and k6 are each an odd number greater than 3.

14. The method of claim 10, wherein S2 is not less than a median of all pixels of the convolved image ordered in ascending pixel value order and/or is not greater than an eighteenth median of all pixels of the convolved image ordered in ascending pixel value order.

15. The method of claim 5, further comprising screening the image for bright spots based on the intensity of the area of the original image where the bright spots are located.

16. The method of claim 10, further comprising screening the image for bright spots based on the intensity of the area of the original image where the bright spots are located.

17. The method of any of claims 2-4, wherein the speckle detection comprises detecting each image in the set of images using a k7 x k8 matrix, comprising,

a k 7-k 8 matrix for determining that a plurality of pixel values in a specified direction are monotonically fluctuating corresponds to one candidate bright spot,

and screening the candidate bright spots by utilizing pixels of at least a part of areas in a corresponding k 7-k 8 matrix to determine the bright spots, wherein k7 and k8 are natural numbers larger than 1, and the k 7-k 8 matrix comprises k 7-k 8 pixels.

18. The method of claim 17, wherein k7 is equal to k8, and/or k7 and k8 are each an odd number greater than 1.

19. The method of any of claims 2-4, further comprising determining subpixel coordinates of the bright spot using a barycentric method.

20. The method of claim 5, further comprising determining subpixel coordinates of the bright spot using a barycentric method.

21. The method of claim 10, further comprising determining subpixel coordinates of the bright spot using a barycentric method.

22. The method of claim 17, further comprising determining subpixel coordinates of the bright spot using a barycentric method.

23. The method of any one of claims 2-4, wherein aligning the set of images comprises using images from an mth round of sequencing for the alignment.

24. The method of claim 5, wherein aligning the set of images comprises using images from an mth round of sequencing for the alignment.

25. The method of claim 10, wherein aligning the set of images comprises using images from an mth round of sequencing for the alignment.

26. The method of claim 17, wherein aligning the set of images comprises using images from an mth round of sequencing for the alignment.

27. The method of claim 23, wherein M is greater than 20.

28. The method of claim 27, wherein M is greater than 50.

29. The method of claim 23, wherein the image sequenced by the mth round comprises a fifth image, a sixth image, a seventh image, and an eighth image, the fifth image, the sixth image, the seventh image, and the eighth image corresponding to the reaction signals of the same species of nucleotide as the first image, the second image, the third image, and the fourth image, respectively,

Converting coordinate systems of a sixth image, a seventh image and an eighth image respectively by taking the coordinate system of the fifth image as a reference, wherein the fifth image and the sixth image are respectively divided into a group of blocks with the size of k9 x k10 in the same way, the k9 and the k10 are natural numbers larger than 30, and the k9 x k10 comprises k9 x k10 pixels;

determining an offset of each block of the sixth image relative to a corresponding block of the fifth image, respectively;

the second image and the first image are aligned based on the offset.

30. The method of any one of claims 24-26, wherein the image sequenced in the Mth round comprises a fifth image, a sixth image, a seventh image, and an eighth image, wherein the fifth image, the sixth image, the seventh image, and the eighth image correspond to the reaction signals of the same nucleotide as the first image, the second image, the third image, and the fourth image, respectively,

converting coordinate systems of a sixth image, a seventh image and an eighth image respectively by taking the coordinate system of the fifth image as a reference, wherein the fifth image and the sixth image are respectively divided into a group of blocks with the size of k9 x k10 in the same way, the k9 and the k10 are natural numbers larger than 30, and the k9 x k10 comprises k9 x k10 pixels;

Determining an offset of each block of the sixth image relative to a corresponding block of the fifth image, respectively;

the second image and the first image are aligned based on the offset.

31. The method of any of claims 2-4, wherein merging the bright spots on the aligned set of images comprises merging a plurality of bright spots within a predetermined range k11 x k12 into one bright spot, where k11 and k12 are natural numbers greater than 1, and k11 x k12 comprises k11 x k12 pixels.

32. The method of claim 5 wherein merging the bright spots on the aligned set of images includes merging a plurality of bright spots within a predetermined range k11 k12 into one bright spot, where k11 and k12 are natural numbers greater than 1, and k11 k12 comprises k11 k12 pixels.

33. The method of claim 10 wherein merging the bright spots on the aligned set of images includes merging a plurality of bright spots within a predetermined range k11 x k12 into one bright spot, where k11 and k12 are natural numbers greater than 1, and k11 x k12 comprises k11 x k12 pixels.

34. The method of claim 17 wherein merging the bright spots on the aligned set of images includes merging a plurality of bright spots within a predetermined range k11 x k12 into one bright spot, where k11 and k12 are natural numbers greater than 1, and k11 x k12 comprises k11 x k12 pixels.

35. The method of any of claims 2-4, wherein the correction of intensity comprises cross-color correction and/or phase correction.

36. The method of claim 5, wherein the correction of intensity comprises cross-color correction and/or phase correction.

37. The method of claim 10, wherein the correction of intensity comprises cross-color correction and/or phase correction.

38. The method of claim 17, wherein the correction of intensity comprises cross-color correction and/or phase correction.

39. The method of claim 35, wherein the image to be inspected is aligned to the set of bright spots corresponding to templates prior to correcting the intensity.

40. The method of claim 39, wherein the correction of intensity comprises a cross color correction based on at least one of images from the same round of sequencing and corresponding to different types of nucleotides.

41. The method of any one of claims 36-38, wherein the correction of intensity comprises a cross color correction based on at least one of images from the same round of sequencing and corresponding to different types of nucleotides.

42. The method of claim 40, wherein image Xi and said image to be inspected are from the same round of sequencing, said image Xi and said image to be inspected correspond to the same field of view, said image to be inspected comprises signals from the nucleotides corresponding to said image Xi, said color correction comprises,

fitting signals of positions of a plurality of corresponding coordinates in a specific area of the image to be detected to obtain a fitting result; and

correcting signals of positions of corresponding coordinates on the image to be detected based on fitting results.

43. The method of claim 41, wherein image Xi and said image to be inspected are from the same round of sequencing, said image Xi and said image to be inspected correspond to the same field of view, said image to be inspected comprises a signal from a nucleotide corresponding to said image Xi, said color correction comprises,

fitting signals of positions of a plurality of corresponding coordinates in a specific area of the image to be detected to obtain a fitting result; and

correcting signals of positions of corresponding coordinates on the image to be detected based on fitting results.

44. The method of claim 42 or 43, wherein the fit is a linear fit.

45. The method of claim 35, wherein the correction of intensity comprises phase correction based on at least one of images from adjacent rounds of sequencing and corresponding to the same kind of nucleotide.

46. The method of any one of claims 36-38, wherein the correction of intensity comprises phase correction based on at least one of images from adjacent rounds of sequencing and corresponding to the same kind of nucleotide.

47. The method of claim 45, wherein the image Yj and the image to be examined are from two adjacent rounds of sequencing, the image Yj and the image to be examined correspond to the same field of view, the image Yj and the image to be examined correspond to the same kind of nucleotide, the phase correction comprises,

fitting signals of positions of a plurality of corresponding coordinates in a specific area of the image to be detected to obtain a fitting result; and

correcting signals of positions of corresponding coordinates on the image to be detected based on fitting relation.

48. The method of claim 46, wherein the image Yj and the image to be examined are from two adjacent rounds of sequencing, the image Yj and the image to be examined correspond to the same field of view, the image Yj and the image to be examined correspond to the same kind of nucleotide, the phase correction comprises,

Fitting signals of positions of a plurality of corresponding coordinates in a specific area of the image to be detected to obtain a fitting result; and

correcting signals of positions of corresponding coordinates on the image to be detected based on fitting relation.

49. A computer readable storage medium storing a program for execution by a computer, the execution of the program comprising performing the method of any one of claims 1-48.

50. A system for identifying one or more bases in a nucleic acid, comprising:

the computer readable storage medium of claim 49; and

at least one processor configured to execute the program stored in the computer-readable storage medium.

51. A system for identifying one or more bases in a nucleic acid, comprising a plurality of modules for performing the steps of the method of any one of claims 1-48.